* Mark Hahn said:
> > way, and is in the 2.2 kernels since ages). For #2, it makes no sence t=
o=20
> > switch to userspace a lot, as userspace effectively would only incremen=
t=20
> > some counter.
>=20
> this is a policy decision, and thus wrong. real webservers want to
> log hits, even on static pages. this, and the fact that most real
> webservers are dominated by non-static pages, is why khttpd shouldn't
> be in the kernel: it's a benchmark special, embarassingly so. our=20
> friends in Redmond will certainly not refrain from pointing out that=20
> the only way Linux can compete is by violating our own abstractions.
Hmm... that seems quite reasonable what you wrote above, but why not to
think about khttpd as something which gives performance boost in some
_special_ cases, be they 5% of the case pool? If it's ok for those 5% and it
doesn't hurt the remaining 95% then why shouldn't it be in the kernel?
=20
marek
--vEao7xgI/oilGqZ+
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iEYEAREBAAYFAjhjgVoACgkQq3909GIf5uovIwCbBPz1GfIckZ3eQGqndFrCnXyZ
L1IAn1g7DMJEUrhsuPRXsZWwugIdkG4e
=9o9o
-----END PGP SIGNATURE-----
--vEao7xgI/oilGqZ+--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/